The objective of the present study was to compare recurrence-free survival (RFS) in early stages (FIGO stages I-II) of epithelial ovarian cancer after adjuvant chemotherapy with carboplatin and a taxane (113 patients) and with carboplatin alone (27 patients). The distribution of clinical and pathological prognostic factors as well as type of primary surgery were comparable in the two groups. Recurrence rate was 21% and RFS was 79% in the series of patients treated with taxane-based chemotherapy and 19% and 81%, respectively, in the series of patients who received single drug carboplatin. Thus, no significant differences were recorded. The major toxicities in the present study were myelosuppression (46%) and neuro-toxicity (26%). Neuro-toxicity was more frequently (P=0.007) recorded and of higher grade (P=0.011) for patients in the carboplatin-taxane series compared with patients in the carboplatin series. RFS for patients in FIGO-stage I was 85% and for patients in FIGO-stage II only 47%. In a multivariate logistic regression analysis of predictive factors for tumor recurrence in the complete series (n=140) the FIGO stage was the only independent and significant (P=0.0006) predictive factor with an odds ratio of 6.4 (95% CI: 2.2-18.9) for stage II versus IA-C. Age, tumor grade and type of adjuvant chemotherapy (+/- taxane) were not significant predictive factors. In the present study, although based on a limited number of patients, we could not find any improvement in recurrence rate or recurrence-free survival for patients treated with a carboplatin-taxane combination regimen compared with patients treated with carboplatin monotherapy. The spectrum of side effects was also in favor of the monotherapy regimen. Further, larger randomized studies are needed to give a final and fully conclusive answer to this question.